WHAT MATTERS IN COLLEGE STUDENT SUCCESS? DETERMINANTS OF COLLEGE RETENTION AND GRADUATION RATES.
Many studies have identified factors of college student retention and graduation rates. This review categorizes the research into three broad categories: institutional factors, student attributes, and financial considerations.
Institutional factors include student/ faculty ratios, student-life programs and services, and specific academic programs such as college-prep, honors courses, or first-year experience classes. Tinto (2006) suggested that such institutional factors encourage students' persistence. University administrators need to know which aspects of these internal investments and institutional management strategies impact student success rates.
Institutional distributions of funding and resources across functional categories indicate a university's priorities and can have significant impact on student outcomes. Hamrick, Schuh, and Shelley (2004) found that instructional and library spending positively impacts student graduation rates. Likewise, Ryan (2004) confirmed that academic and instructional spending improved graduation rates. Webber and Ehrenberg (2010) extended their analyses to include student support services and instruction expenditures, and found that both increase graduation and retention rates. A university's specific purchases matter as well as the categories of expenditures. For instance, spending on tenured and tenure-track faculty instead of nontenure-track instructors positively impacts graduation rates (Ehrenberg & Zhang, 2005).
Instead of looking at broad expenditure categories, other researchers have investigated the effectiveness of specific programs targeting student populations. For example, programs can create a shared university experience for students, such as first-year students (Colton, Connor, Shultz, & Easter, 1999; Harrington et al., 2016; Inkelas, Daver, Vogt, & Leonard, 2007; Noble, Flynn, Lee, & Hilton, 2007) or students within a common major (Dagley, Georgiopoulos, Reece, & Young, 2015; Watterson & Carnegie, 2011). Other programs target students with specific demographic attributes such as students of color (Aragon & Rios Perez, 2006), nontraditional students (Wyatt, 2011) and first-generation students (Inkelas et al., 2007).
Consistent with Tinto's (1987) seminal paper, studies have shown that common courses or shared experiences can improve student success by integrating students into the university community. In particular, learning communities or first-year experience courses have had positive impacts on retention rates and grade point averages (GPAs; Burgette & Magun-Jackson, 2008; Clark & Cundiff, 2011; Jamelske, 2009; Miller, Janz, & Chen, 2007; Yockey & George, 1998) and graduation rates (Lang, 2007; Noble et al., 2007). Some universities have additionally integrated students into the university community through mandatory on-campus residence. Studies comparing student success measures such as GPA and retention across living arrangements have shown that students who lived at home and students who lived in residence halls had success rates that were consistently higher than those of students who lived in fraternity or sorority housing and were, in some cases, higher than those of students who lived in off-campus apartments (Blimling, 1989; Bowman & Partin, 1993; Pascarella, 1984, 1985).
Involvement and engagement have been identified as keys to student success in college. Students who feel connected to their academic endeavors are more likely to succeed (Allen, Robbins, Casillas, & Oh, 2008; Baker & Robnett, 2012; Hunt, Boyd, Gast, Mitchell, & Wilson, 2012; Morrow & Ackermann, 2012; Svanum & Bigatti, 2009). Attention to the quality of the classroom experience for students is one academic condition that promotes student engagement.
An obvious prerequisite to reaping the benefit of classroom interactions, especially in the first year of college, is to ensure that students participate in class (Knaggs, Sondergeld, & Schardt, 2015). Some universities track class attendance among first-year students to increase student accountability (Lotkowski, Robbins, & Noeth, 2004; Hassel & Lourey, 2005) and academic success (Crede, Roch, & Kieszczynka, 2010). Cartney and Rouse (2006) recommended facilitating small-group learning opportunities to improve student first-year success. Some universities limit class size to encourage participation and accountability. Studies have found that a large class size can adversely impact student satisfaction and student evaluations of instructors (DeShields, Kara, & Kaynak, 2005; Miles & House, 2015). Other studies have tracked an inverse relationship between class size and student performance in a specific class (De Paola et al., 2013; Morris & Scott, 2014). Chapman and Ludlow (2010) confirmed that a large class size adversely impacts graduate and undergraduate perceptions of learning. Diette and Raghav (2015) found that students' grades were adversely affected by class size at a private, highly selective liberal arts college, and that this effect was greater for first-year students and students with lower SAT scores. The average class size in the Diette and Raghav study was 20.2, which may not be generalizable to larger public institutions.
Another category of factors that impact students' success in college includes individual attributes, such as behaviors, motivation, academic preparation, demographic factors, and family characteristics, such as parents' and siblings' degree attainment. Students who are more academically prepared, not surprisingly, have more success in college (Braunstein & McGrath, 1997; Johnson, 2008; Kirby & Sharpe, 2001). However, regardless of preparation, behaviors and decisions while in school impact students' achievement. Modfidi, Amani, and Brown (2014) found that trait gratitude and grateful coping strategies improved student success along several dimensions, including GPA and college persistence. Their survey included only 54 students; similar results were found by Slanger, Berg, Fisk, and Hanson (2015) using ten years of College Student Inventory data. They identified motivational factors impacting students' success, including GPA and retention. DeBerard, Spielmans, and Julka (2004) documented that better coping strategies, healthy choices regarding smoking and drinking, and social/parental support positively affect students' academic performance.
Financial constraints can also inhibit student performance. In some cases, universities and policy makers distribute financial aid, typically in the form of scholarships, grants, and/or loans. Institutional aid or scholarships are usually distributed to students based on academic merit, and in some cases for athletics. Government-supported aid, grants, and loans are typically need-based. In an extensive review of higher education literature, Pascarella and Terenzini (2005) reported that financial aid was beneficial for student persistence and degree completion, particularly for those in need. In distinguishing between the different types of aid, mixed patterns emerge. Generally, grant and scholarship aid had positive effects on retention and graduation rates, while loans had either a positive impact or no effect (Pascarella & Terenzini, 2005).
Glocker (2011) found that student aid increased the probability of graduation and decreased the duration of studies for students in Germany. In a U.S. study, Wohlgemuth et al. (2007) and Whalen, Sanders, and Shelley (2009) found that financial aid increased retention and graduation rates among incoming freshmen. In Singell's (2004) study of students at the University of Oregon, need- and merit-based aid increased retention overall; however, need-based aid actually decreased graduation rates. He also surveyed unretained students who indicated their decisions to leave were largely financial.
When using research to identify practical strategies for institutional investments, case studies may be too narrow in scope to be generalizable to other contexts, and macro-level analyses may not provide sufficient specificity to create institution-specific strategies. Moreover, some of the contributing factors to student success may not be under the control of college and university administrators (Habley et al., 2012). This study attempts to address these dilemmas by simultaneously evaluating institutional factors (such as programs specifically designed to improve student retention, class size, and on-campus/residential living), student attributes (such as academic preparation, first-year performance, and demographic characteristics), and financial considerations (such as grant aid, scholarships, and subsidized and unsubsidized student loans).
One contribution of this paper is to provide an example of using institutional data across multiple dimensions to quantify contributing factors to institutional metrics of concern. Excluding relevant factors, such as financial aid or academic experiences, can yield misspecified models and potentially translate into universities engaging in inefficient or even ineffective investment strategies. Another contribution of this study is the inclusion of class size, using individual student transcript records, matched with financial aid and admission data to track the retention and graduation rates for entire incoming freshmen classes. Our data includes nearly 13,000 incoming freshmen over a seven-year period at one midsized public university in the southeastern United States.
Detailed student records were collected for first-time, full-time freshmen whose first enrollment occurred during the period 1998-2004. The initial observation year, 1998, was the first year the university adopted its online data management software system. To better isolate the effects of living on-campus, we used an end date of 2004 because it preceded the on-campus residence requirement for all incoming freshmen. The primary student records came from the university's database and contained the students' admission records, including demographics, ACT scores, financial aid, and first-year experiences such as classes and grades. The university also implemented an attendance program in 1998 that centrally collected absenteeism records for freshmen. If students missed classes, program staff would contact the students directly and offer interventions and support. Absenteeism and on- or off-campus residential status records collected by this program were matched with the official university records to create our data set.
Descriptive statistics are presented in Table 1. There were nearly 13,000 first-time freshmen in this sample. About half were male, and three-quarters were White. Most of the students were in-state residents (72%). In terms of their preparation for college, the average high school GPA for core classes was 3.13, and the average ACT composite score was 23.4. In the first year, 16% of the students had chronic absenteeism in at least one class (more than five absences reported). The average GPA in general education courses in the first year was 2.56, and the average class size was 82.2 students. During this period, although living on campus was optional, 87% of freshmen chose to live in residence halls on campus.
Students' success can depend on their own preparation and effort, class experiences, living arrangements, and financial status. Thus, we also included metrics about students' financial support in our models: 64% of the freshmen received grant aid, 34% received merit-based scholarships, 3% received athletic scholarships, and 27% utilized subsidized loans. This study focused on student success as measured by retention and graduation rates. Of the 12,812 incoming students, 80% were retained, meaning they were enrolled in the following fall semester. Fifty-seven percent of the freshmen graduated within 6 years.
Analyses and Results
The first hurdle in academic achievement is remaining in school. Thus, our initial measure of student success for incoming freshmen was retention. Several factors can impact freshman-to-sophomore retention rates, including institutional conditions, demographics, socioeconomic status, student behaviors, and academic ability and performance. We estimated a limited dependent model using Probit analysis to estimate the probability that a first-time, full-time freshman was retained into the fall semester following his or her initial fall enrollment. We included the marginal effects to identify the explanatory power of each independent variable.
We included demographic controls for sex, race, and age. There were no consistent a priori theoretical expectations about the effects of these attributes. However, we included a geographic variable that we expected to be positive. We posited that students who were in-state residents may be more likely to persist in enrollment compared to nonresident students, in part due to their proximity to home and the lower price of resident tuition. We expected that students who lived in residence halls would be more integrated in the university and thus more likely to be retained.
To control for academic ability, the model includes high school core GPAs and ACT scores. Although both are measures of academic attributes, they measure different characteristics. GPA is a measure of academic performance, the degree to which the individual can apply knowledge and perform tasks assessed for grades. ACT is a measure of college readiness. Both measures were expected to positively impact retention rates.
In 1998, this university embarked on a program that actively tracked freshmen class attendance and had faculty report absences to program administrators. Due to the reliance on voluntary faculty reporting, there may be systematic errors in the data if individual faculty members consistently failed to report or did not report absences accurately. Thus, we included a measure of chronic absenteeism that measured whether a student had more than five absences reported for any one class. Persistently missing class would presumably undermine a student's probability of being retained.
The other academic metrics of first-year experiences are consistently based on general education course work because those classes are the most homogeneous across the student population. For example, introductory engineering courses may have different class sizes, content, and rigor compared to introductory classes for majors in business, social sciences, or the humanities. At this university, general education requirements are the same across all majors and include the same core course options. This common course work structure provided a basis for comparison.
Reviewing individual student transcripts for general education courses was not a straightforward endeavor. Several class options fulfilled the general education requirements in each category: English, math, natural science, social science, humanities, and fine arts. The matches between category and courses were not complicated. However, identifying which classes on each student's transcript counted toward the fulfillment of the general education requirements was more complicated. For example, a social science major may take more than the minimum 6 hours of social science courses within the first year. We had to identify which social science courses to include in the general education GPA and class size calculations. Thus, we developed a series of limiting decision criteria. We decided to first count courses by when the course was taken, excluding any Advanced Placement (AP) courses or College-Level Examination Program (CLEP) credits. Courses were included sequentially up the point of general education category fulfillment (i.e., 6 hours of social sciences). If a student exceeded the general education hours required within one of the broad categories by taking multiple courses within the same semester, we had to select which course grade and class size to count. Thus, the next limiter was the level of the course, with freshman-level courses included before sophomore-level courses. If multiple freshman-level social science courses were taken during one semester, the next limiter was to include the course in which the student earned the highest grade. When semester, course level, and grade earned were equivalent, the final limiter was enrollment; the lowest-sized class was included. For example, a student might take one social science course in his or her first semester, which would be counted in the general education metrics (class size and GPA) for this study. If the student then took two more social science courses in the second term, a freshman-level course would be included before a sophomore-level course. If both courses were sophomore-level social science classes, then the grade earned would matter. If the student earned A's in both courses, the class with the lower enrollment would count as the student's final social science general education course when we calculated the GPA variable and average class size. This process was repeated for each of the 12,812 students for all six categories of general education requirements. Based on these specifications, GPA may be upwardly biased, and the class size may be downwardly biased.
Our a priori expectations were that students in smaller classes would receive more individualized attention and feel more connected to their academic experience, and were therefore more likely to be retained. We expected the class size coefficient to be negative. Students' performances in those courses could also influence the students' probability of persisting into the following fall semester; therefore, higher average GPA should increase the probability of retention. The freshman GPA coefficient should be positive.
Students' financial statuses may impact their ability to persist in enrollment. Students who were the most financially needy were eligible to receive federal grants. Because grants reduce financial burdens and do not require payback, we assumed they increased the probability of being retained. Students may also borrow money. Students who were eligible would first opt to take subsidized loans, which defer interest accumulation. Unsubsidized loans go either to low-income students who have borrowed their maximum subsidized loans or to higher-income students who were eligible for loans but not for the interest subsidy. Because loans only defer the cost of education, they may not influence the probability of retention. Students may also have received scholarships. Institutional scholarships at this university are based on academic merit or athletics; we posited that both would increase the probability of being retained into sophomore year.
The model for these analyses included demographics, academic preparation, financial aid, absenteeism across all classes, and first-year academic experiences (such as average class size and GPAs in general education courses). In the first model, the probability of retention from the first fall term to the following fall term was estimated as a binary dependent variable:
Retention = [[alpha].sub.0] + [[alpha].sub.1]HS GPA + [[alpha].sub.2]ACT + [[alpha].sub.3]MALE + [[alpha].sub.4]WHITE + [[alpha].sub.5]AGE+ [[alpha].sub.6]RES + [[alpha].sub.7]ABSENT + [[alpha].sub.8]FR_GPA + [[alpha].sub.9]CLASS_SIZE + [[alpha].sub.10]DORM + [[alpha].sub.11]GRANT + a[[alpha].sub.12]SUB_ LOAN + [[alpha].sub.13]UNSUB_LOAN + [[alpha].sub.14]SPORT + [[epsilon].sub.1]. (1)
During the period of our study, merit-based scholarships at this university depended almost exclusively on the student's ACT score; therefore, the scholarship variable was excluded from this model specification. Including the ACT score is the preferred specification because this score has ordinal values, whereas the merit scholarship variable is binary.
If high school GPA and ACT are both measures of a student's academic preparedness, then the model may be misspecified. Thus, Equation 2 was estimated excluding the ACT score. This exclusion allowed us to include merit-based scholarships.
Retention = [[alpha].sub.0] + [[alpha].sub.1]HS GPA + [[alpha].sub.2]MALE + [[alpha].sub.3]WHITE + [[alpha].sub.4]AGE+ [[alpha].sub.5]RES + [[alpha].sub.6]ABSENT + [[alpha].sub.7]FR_GPA + [[alpha].sub.8]CLASS_SIZE + [[alpha].sub.6]DORM + [[alpha].sub.10]GRANT + [[alpha].sub.11]SUB_ LOAN + [[alpha].sub.12]UNSUB_LOAN + [[alpha].sub.13] SCHOLARSHIP + [[alpha].sub.14]SPORT + [[epsilon].sub.2]. (2)
The results of these estimations are presented in Table 2. The first column of each model specification indicates the sign and significance of the independent variable. The second column indicates the magnitude or marginal impact of each variable. In both equations, high school GPA positively influenced the probability of being retained. Each additional GPA point increased the probability of retention by 0.01%. From Equation I, each additional point earned on the ACT increased the retention probability by 0.1%. There were no differences between White and non-White students or between male or female students in terms of retention from freshman fall semester to the following fall term. However, in Equation 2, the age variable is positive and significant; each additional year of age of the incoming freshman increased the probability of retention by 0.6%. Higher grades in general education courses during freshman year increased the probability of returning the following fall by 0.07% for each additional point on the 4-point scale. Somewhat unexpectedly, in-state residency, absenteeism, and living on campus did not impact the probability of being retained. However, larger classes adversely impacted the probability of returning. As the average class size increased by 10 students, the probability of retention fell by 0.3% (equation 1) and by 0.2% (equation 2). As expected, grants increased the probability of retention. This was a binary variable, so receipt of a grant of any amount increased the probability of retention by 7.7% in the first specification and 6.9% in the second specification.
Scholarships also increased the probability of retention. Receiving a merit-based scholarship increased the probability of success by 12.5%. Scholarship athletes were approximately 10% more likely to be retained ceteris paribus (9.4% in Equation 1 and 11.3% in Equation 2).
In summary, retention rates were higher for academically prepared students, those who performed well in classes, and those who received merit-based and athletic scholarships and grant aid. These results are consistent with other research findings and a priori expectations. In addition, we found that smaller class sizes improved retention rates. Retention is the first step toward achieving academic success, considered here as graduating within 6 years. In the following section, similar equations are estimated using 6-year graduation rates as the dependent variable.
Similar factors impacted graduation rates as those impacting retention: demographics, academic experiences, and financial factors. To control for academic ability, the model included high school core GPAs and ACT scores. Both were expected to positively impact graduation rates. Race, sex, and age were included as demographic controls. In-state students were expected to be more likely to persist until graduation. Data on chronic absenteeism were reported only in the freshman year; however, absenteeism can still serve as a behavioral proxy that may impact graduation rates. We also included on-campus residence in the freshman year as a potential indicator of campus engagement, as this experience may impact graduation rates. We included institutional factors such as class size, academic performance while at the university, and financial status in the forms of grants, scholarships, and loans.
Using Probit, a linear dependent regression analysis that assumes a normal distribution of the error terms, the following equation was estimated:
6-Year Graduate Rate = [[alpha].sub.0] + [[alpha].sub.1]HS GPA + [[alpha].sub.2] ACT+ [[alpha].sub.3]MALE + [[alpha].sub.4]WHITE + [[alpha].sub.5]AGE+ [[alpha].sub.6]RES + [[alpha].sub.7]ABSENT + [[alpha].sub.8]FR_GPA + [[alpha].sub.9]CLASS_SIZE + [[alpha].sub.10]DORM + [[alpha].sub.11]GRANT + [[alpha].sub.12]SUB_LOAN + [[alpha].sub.13]UNSUB_LOAN + [[alpha].sub.14]SPORT + [[epsilon].sub.3] (3)
We modified Equation 3 to substitute a binary merit-based scholarship variable in place of the ACT composite variable as we did in Equation 2, yielding the specification in Equation 4:
6-YearGraduateRate = [[alpha].sub.0] + [[alpha].sub.1]HSGPA + [[alpha].sub.2]MALE + [[alpha].sub.3]WHITE + [[alpha].sub.4]AGE+ [[alpha].sub.5]RES + [[alpha].sub.6]ABSENT + [[alpha].sub.7]FR_GPA + [[alpha].sub.8]CLASS_SIZE + [[alpha].sub.9]DORM + [[alpha].sub.10]GRANT + [[alpha].sub.11]SUB_LOAN + [[alpha].sub.12]UNSUB_ LOAN + [[alpha].sub.13]SCHOLARSHIP + [[alpha].sub.14]SPORT + [[epsilon].sub.4]. (4)
The results of these estimations are presented in Table 3. High school GPA did not contribute significantly to college graduation rates, although each point higher on the ACT increased the probability of graduating by 0.5%. There were no differences by sex or race. Older students were less likely to complete their degrees within 6 years. Each additional year of age reduced the probability of graduation by 1.9%. It may be that older students did not complete their degrees or did not complete their degrees within 6 years of initial enrollment. Several factors that researchers have indicated as important in explaining student success were not statistically different from zero in our model, including in-state residency, chronic absenteeism in freshman year, and living in a residence hall. Things that persistently mattered were either academic or financial. Higher freshman GPAs increased the probability of graduating; each additional grade point in general education courses in the first year on the 4-point scale led to a 0.3% higher likelihood of graduating. Smaller class sizes helped, too; when freshmen had classes that were, on average, 10 students smaller, the students were 0.4% more likely to graduate. However, the largest impact on graduation was financial.
Merit-based scholarships increased the probability of graduating by 18.4%. There was likely a selection bias. The students who were meritorious were inherently more likely to graduate, in large part due to the same abilities and motivation that earned them the merit-based scholarships. We do not claim that the scholarship itself increased graduation rates, but that the meritorious attributes rewarded by the scholarship contributed to higher graduation rates. In addition, the scholarship reduced financial burden, which contributed to academic persistence through graduation. However, athletic scholarships did not impact graduation rates. There was no significant difference in the graduation rates of students with athletic scholarships and their peers. Although athletic scholarships positively impacted retention rates (Table 2), the impact on student success did not persist through to graduation.
Grant aid increased the probability of graduating by about 9% (9.9% in the first specification and 8.9% in the second specification). Receiving student loans decreased the probability of graduation. Receiving a subsidized loan decreased the probability of graduating by 19% in the first specification and 15.1% in the second specification. In addition, unsubsidized loans decreased the probability of graduating by 2.5%. Clearly, financial aid was a large and significant factor in graduation rates, but it is important to note that the types of aid had very different effects--grants were positive and loans were negative.
This study contributes to the literature on college retention and graduation by demonstrating an application of institutional analysis that combines data from several different sources across a single university. This study systematically tracked individual student transcripts to capture the marginal impacts of freshman general education academic experiences on retention and graduation rates.
The practical implications of this work suggest that universities should invest in smaller class sizes and focus on students' financial constraints to improve student success. Surprisingly, two factors typically considered important for freshmen retention and eventual graduation, absenteeism and on-campus residence, were not found to be significant in our models.
Although the nature of the individual-level analysis justifies use of a single institution, the tradeoff is that the results may not be sufficiently generalizable. This study provides an example of combining information across institutional datasets to inform strategic decision making. Future work could apply a similar methodology to other universities' student populations using more current data and additional factors, such as access to online education. Researchers can also modify the metrics of student success to align with specific institutional goals, funding models, or administrators' focus areas.
Dr. Meghan Millea
East Carolina University
Allen, J., Robbins, S. B., Casillas, A., & Oh, I. S. (2008). Third-year college retention and transfer: Effects of academic performance, motivation, and social connectedness. Research in Higher Education, 49(1), 647-664.
Aragon, S. R., & Rios Perez, M. (2006). Increasing retention and success of students of color at research-extensive universities. New Directions for Student Services, 2006(114), 81-91.
Baker, C. N., & Robnett, B. (2012). Race, social support and college student retention: A case study. Journal of College Student Development, 53(2), 325-335.
Blimling, G. S. (1989). A meta-analysis of the influence of college residence halls on academic performance. Journal of College Student Development, 30(4), 298-308.
Bowman, R. L., & Partin, K. E. (1993). The relationship between living in residence halls and academic achievement. College Student Affairs Journal, 73(1), 71-78.
Braunstein, A., & McGrath, M. (1997). The role of economic factors in higher education persistence. International Advances in Economic Research, 3(3), 325-326.
Braxton, J. M., Hartley, III, H. V., & Lyken-Segosebe, D. (2014). Students at risk in residential and commuter colleges and universities. In D. Hossler & B. Bontrager (Eds.), Handbook of strategic enrollment management (pp. 289). San Francisco, CA: Jossey-Bass.
Burgette, J. E., & Magun-Jackson, S. (2008). Freshman orientation, persistence, and achievement: A longitudinal analysis. Journal of College Student Retention: Research, Theory & Practice, 10(3), 235-263.
Cartney, P., & Rouse, A. (2006). The emotional impact of learning in small groups: Highlighting the impact on student progression and retention. Teaching in Higher Education, 11(1), 79-91.
Chapman, L., & Ludlow, L. (2010). Can downsizing college class sizes augment student outcomes? An investigation of the effects of class size on student learning. Journal of General Education, 59(2), 105-123.
Clark, M. H., & Cundiff, N. L. (2011). Assessing the effectiveness of a college freshman seminar using propensity score adjustments. Research in Higher Education, 52(6), 616-639.
Colton, G. M., Connor, U. J., Shultz, E. L., & Easter, L. M. (1999). Fighting attrition: One freshman year program that targets academic progress and retention for at-risk students. Journal of College Student Retention: Research, Theory & Practice, 1(2), 147-162.
Crede, M., Roch, S. G., & Kieszczynka, U. M. (2010). Class attendance in college: A meta-analytic review of the relationship of class attendance with grades and student characteristics. Review of Educational Research, 80(2), 272-295.
Dagley, M., Georgiopoulos, M., Reece, A., & Young, C. (2015). Increasing retention and graduation rates through a STEM learning community. Journal of College Student Retention: Research, Theory & Practice, 18(2), 167-182.
DeBerard, M. S., Spielmans, G. I., & Julka, D. L. (2004). Predictors of academic achievement and retention among college freshmen: A longitudinal study. College Student Journal, 38(1), 66.
De Paola, M., Ponzo, M., & Scoppa, V. (2013). Class size effects on student achievement: Heterogeneity across abilities and fields. Education Economics, 21(2), 135-153.
DeShields, Jr., O. W., Kara, A., & Kaynak, E. (2005). Determinants of business student satisfaction and retention in higher education: Applying Herzberg's two-factor theory. International Journal of Educational Management, 19(2), 128-139.
Diette, T. M., & Raghav, M. (2015). Class size matters: Heterogeneous effects of larger classes on college student learning. Eastern Economic Journal, 41(2), 273-283.
Ehrenberg, R. G., & Zhang, L. (2005). Do tenured and tenure-track faculty matter? Journal of Human Resources, 40(3), 647-659.
Glocker, D. (2011). The effect of student aid on the duration of study. Economics of Education Review, 30(1), 177-190.
Habley, W. R., Bloom, J. L., & Robbins, S. (2012). Increasing persistence: Research-based strategies for college student success. San Francisco, CA: Jossey-Bass.
Hamrick, F. A., Schuh, J. H., & Shelley, II, M. C. (2004). Predicting higher education graduation rates from institutional characteristics and resource allocation. Education Policy Analysis Archives, 72(19), 1.
Harrington, M. A., Lloyd, A., Smolinski, T. & Shahin, M. (2016). Closing the gap: First year success in college mathematics at an HBCU. Journal of The Scholarship of Teaching and Learning, 16(5), 92-106.
Hassel, H., & Lourey, J. (2005). The dea(r)th of student responsibility. College Teaching, 33(1), 2-13.
Hunt, P. F., Boyd, V. S., Gast, L. K., Mitchell, A., & Wilson, W. (2012). Why some students leave college during their senior year. Journal of College Student Development, 33(5), 737-742.
Inkelas, K. K., Daver, Z. E., Vogt, K. E., & Leonard, J. B. (2007). Living-learning programs and first-generation college students' academic and social transition to college. Research in Higher Education, 48(4), 403-434.
Jamelske, E. (2009). Measuring the impact of a university first-year experience program on student GPA and retention. Higher Education, 57(3), 373-391.
Johnson, I. (2008). Enrollment, persistence and graduation of in-state students at a public research university: Does high school matter? Research in Higher Education, 49(8), 776-793.
Kirby, D., & Sharpe, D. (2001). Student attrition from Newfoundland and Labrador's public college. Alberta Journal of Educational Research, 47(4), 353.
Knaggs, C. M., Sondergeld, T. A., & Schardt, B. (2015). Overcoming barriers to college enrollment, persistence, and perceptions for urban high school students in a college preparatory program. Journal of Mixed Methods Research, 9(1), 7-30.
Lang, D. (2007). The impact of a first-year experience course on the academic performance, persistence, and graduation rates of first-semester college students at a public research university. Journal of the First-Year Experience & Students in Transition, 79(1), 9-25.
Lotkowski, V. A., Robbins, S. B., & Noeth, R. J. (2004). The role of academic and non-academic factors in improving college retention. Iowa City, 1A: American College Testing.
Miles, P., & House, D. (2015). The tail wagging the dog: An overdue examination of student teaching evaluations. International Journal of Higher Education, 4(2), 116.
Miller, J., Janz, J., & Chen, C. (2007). The retention impact of a first-year seminar on students with varying pre-college academic performance. Journal of the First-Year Experience & Students in Transition, 79(1), 47-62.
Mofidi, T., El-Alayli, A., & Brown, A. A. (2014). Trait gratitude and grateful coping as they related to collect student persistence, success, and integration in school. Journal of College Student Retention: Research. Theory & Practice, 163), 325-349.
Morris, Sr., D. E., & Scott, J. (2014). A revised pilot study examining the effects of the timing and size of classes on student performance in introductory accounting classes. Research in Higher Education Journal, 23, 1.
Morrow, J., & Ackermann, M. (2012). Intention to persist and retention of first-year students: The importance of motivation and sense of belonging. College Student Journal, 46(3), 483-491.
Noble, K., Flynn, N. T., Lee, J. D., & Hilton, D. (2007). Predicting successful college experiences: Evidence from a first-year retention program. Journal of College Student Retention: Research, Theory & Practice, 9(1), 39-60.
Pascarella, E. T. (1984). College environmental influences on students' educational aspirations. Journal of Higher Education, 55(6), 751-771.
Pascarella, E. T. (1985). Students' affective development within the college environment. Journal of Higher Education, 5(5(6), 640-663.
Pascarella, E. T., & Terenzini, P. T. (2005). How college affects students: A third decade of research. San Francisco, CA: Jossey-Bass.
Ryan, J. F. (2004). The relationship between institutional expenditures and degree attainment at baccalaureate colleges. Research in Higher Education, 45(2), 97-114.
Singell, L. D. (2004). Come and stay a while: Does financial aid effect retention conditioned on enrollment at a large public university? Economics of Education Review, 23(5), 459-471.
Slanger, W. D., Berg, E. A., Fisk, P. S., & Hanson, M. G. (2015). A longitudinal cohort study of student motivational factors related to academic success and retention using the College Student Inventory. Journal of College Student Retention: Research, Theory & Practice, 17(3), 278-302
Svanum, S., & Bigatti, S. M. (2009). Academic course engagement during one semester forecasts college success: Engaged students are more likely to earn a degree, do it faster, and do it better. Journal of College Student Development, 50(1), 120-132.
Tinto, V. (1987). Leaving college: Rethinking the causes and cures of student attrition. Chicago, IL. University of Chicago Press.
Tinto, V. (2006). Research and practice of student retention: What next? Journal of College Student Retention: Research, Theory & Practice, S(1), 1-19.
Watterson, C. A., & Carnegie, D. A. (2011). Increasing student retention and success: Survey results and the success of initiatives to create an engineering student community. In: Global Engineering Education Conference (EDUCON), 2011 IEEE (pp. 191-200). Amman, Jordan: IEEE.
Webber, D. A., & Ehrenberg, R. G. (2010). Do expenditures other than instructional expenditures affect graduation and persistence rates in American higher education? Economics of Education Review, 29(6), 947-958.
Whalen, D., Saunders, K., & Shelley, M. (2009). Leveraging what we know to enhance short-term and long-term retention of university students. Journal of College Student Retention: Research, Theory & Practice, 11(3), 407-430.
Wohlgemuth, D., Whalen, D., Sullivan, J., Nading, C., Shelley, M., & Wang, Y. R. (2007). Financial, academic, and environmental influences on the retention and graduation of students. Journal of College Student Retention: Research, Theory & Practice, S(4), 457-175.
Wyatt, L. G. (2011). Nontraditional student engagement: Increasing adult student success and retention. Journal of Continuing Higher Education, 59(1), 10-20.
Yockey, F., & George, A. (1998). The effects of a freshman seminar paired with supplemental instruction. Journal of the First-Year Experience & Students in Transition, 10(2), 57-76.
Table 1: Descriptive Statistics for Incoming Freshmen 1998-2004 Mean (SD) 6-Year = 1 if the student received degree 0.573 Graduation Rate by the 6th year; (0.495) = 0 otherwise" HSGPA GPA from high school core classes 3.132 based on a 4-point scale (0.597) ACT Composite ACT score 23.42 (4.540) MALE = 1 if the student is male 0.449 = 0 otherwise (0.497) WHITE = 1 if the student identified 0.752 ethnicity as Caucasian (0.432) = 0 otherwise AGE Age calculated as of August 15 of 18.056 admission year using birthdate (0.909) recorded in admission file. In-State Resident = 1 if the student identified 0.720 (RES) as in-state resident (0.449) = 0 otherwise Chronic = 1 if the student had more 0.164 Absenteeism than 5 absences in 1 class (0.370) (ABSENT) = 0 otherwise Freshman General GPA average for general education 2.56 Education GPA courses taken in the first year (0.992) (FRGPA) CLASS SIZE Average class size in general 82.187 education courses taken the first (29.665) year maximum value = 312, minimum value = 13 Residence Hall = 1 if the student lived in 0.865 (DORM) a residence hall on campus (0.342) = 0 otherwise Received Grant = 1 if the student received 0.640 (GRANT) financial aid grant (0.480) = 0 otherwise Subsidized Loan = 1 if the student received and 0.267 (SUB LOAN) took out a subsidized loan (0.442) = 0 otherwise Unsubsidized Loan = 1 if the student received and 0.171 (UNSUBLOAN) look out an unsubsidized loan (0.376) = 0 otherwise Merit Scholarship = 1 if the student received a 0.342 (SCHOLARSHIP) merit-based institutional scholarship (0.474) = 0 otherwise Athletic = 1 if the student received an 0.026 Scholarship athletic scholarship (0.161) (SPORT) = 0 otherwise N = 12,812 (a) The 0 value for the 6-year graduate rate includes students who were not retained into the 6th year and students who had not completed their degrees but were still enrolled in their undergraduate program in the 6th year. Table 2: Probit Results for Retention (Freshman Fall to Following Fall) Estimation (1998-2004) Equation 1 Coefficient Marginal (t statistic) effects Constant 0.372 0.103 (1.425) HSGPA 0.0003 * 0.0001 * (2.452) ACT 0.004 * 0.001 * (4.434) MALE -0.005 -0.001 (-0.196) WHITE -0.0001 -0.0000 (-0.497) AGE 0.022 0.006 (1.582) In-State Resident 0.0000 0.0000 (RES) (0.630) Chronic Absenteeism (ABSENT) -0.0000 -0.0000 (-0.209) Freshman General Education GPA 0.003 * 0.0007 * (FRGPA) (14.979) CLASSSIZE -0.001 * -0.0003 * (-4.940) Residence Hall (DORM) -0.0000 -0.0000 (-0.082) Received Grant (GRANT) 0.269 * 0.077 * (9.571) Subsidized Loan (SUB LOAN) -0.288 * -0.085 * (-9.664) Unsubsidized Loan -0.002 -0.001 (UNSUBLOAN) (-0.065) Merit Scholarship (SCHOLARSHIP) Athletic Scholarship 0.407 * 0.094 * (SPORT) (4.356) Number of observations 12,812 Equation 2 Coefficient Marginal (t statistic) effects Constant 0.274 0.075 (1.078) HSGPA 0.0003 * 0.0001 * (2.268) ACT MALE 0.027 0.007 (0.337) WHITE -0.0001 -0.0000 (-0.550) AGE 0.024 ** 0.006 ** (1.756) In-State Resident 0.0000 0.0000 (RES) (0.753) Chronic Absenteeism (ABSENT) -0.0000 -0.0000 (0.096) Freshman General Education GPA 0.0026 * 0.0007 * (FRGPA) (14.764) CLASSSIZE -0.001 * -0.0002 * (-4.955) Residence Hall (DORM) -0.0000 -0.0000 (-0.766) Received Grant (GRANT) 0.247 * 0.069 * (8.692) Subsidized Loan (SUB LOAN) -0.193 * -0.055 * (-6.292) Unsubsidized Loan -0.0103 -0.003 (UNSUBLOAN) (-0.299) Merit Scholarship 0.494 * 0.125 * (SCHOLARSHIP) (16.188) Athletic Scholarship 0.532 * 0.113 * (SPORT) (5.677) Number of observations * p < 0.05 ** p < 0.10 Table 3: Probit Results for 6-Year Graduation Estimation (1998-2004) Equation 3 Coefficient Marginal (t statistic) effects Constant 1.055 * 0.4155 * (3.473) HSGPA 0.0002 0.0001 (1.823) ACT 0.005 * 0.0020 * (5.335) MALE 0.005 0.0020 (0.217) WHITE -0.0001 -0.0003 (-0.447) AGE -0.049 * -0.0191 * (-3.029) In-State Resident -0.0001 -0.0000 (RES) (-0.418) Chronic Absenteeism (ABSENT) 0.0000 0.0000 (0.036) Freshman General Education GPA (FR_GPA) 0.003 * 0.0012 * (8.427) CLASSS1ZE -0.001 * -0.0004 * (-3.757) Residence Hall (DORM) -0.0000 -0.0000 (-0.082) Received Grant (GRANT) 0.251 * 0.0989 * (10.095) Subsidized Loan (SUB LOAN) -0.480 * -0.1895 * (-17.971) Unsubsidized Loan (UNSUB LOAN) -0.061 * -0.0243 * (-2.044) Merit Scholarship (SCHOLARSHIP) Athletic Scholarship -0.087 -0.0345 (SPORT) (-1.232) Number of observations 12,812 Equation 4 Coefficient Marginal (t statistic) effects Constant 0.9734 * 0.3826 * (3.299) HSGPA 0.0002 0.0010 (1.704) ACT MALE 0.0079 0.0031 (0.338) WHITE -0.0009 -0.0003 (-0.451) AGE -0.048 * -0.0188 * (-3.018) In-State Resident -0.0000 -0.000 (RES) (-0.256) Chronic Absenteeism (ABSENT) 0.0000 0.0000 (0.163) Freshman General Education GPA (FR_GPA) 0.0029 * 0.0011 * (8.518) CLASSS1ZE -0.0010 * -0.0004 * (-3.310) Residence Hall (DORM) -0.0001 -0.0000 (-0.308) Received Grant (GRANT) 0.2267 * 0.0894 * (9.037) Subsidized Loan (SUB LOAN) -0.3821 * -0.1510 * (-13.963) Unsubsidized Loan (UNSUB LOAN) -0.0677 * -0.0267 * (-2.227) Merit Scholarship (SCHOLARSHIP) 0.4793 * 0.1838 * (18.868) Athletic Scholarship 0.0485 0.0190 (SPORT) (0.681) Number of observations * p < 0.05 ** p < 0.10
|Printer friendly Cite/link Email Feedback|
|Author:||Millea, Meghan; Wills, R.; Elder, A.; Molina, D.|
|Date:||Jun 22, 2018|
|Previous Article:||GRADUATION 101: CRITICAL STRATEGIES FOR AFRICAN AMERICAN MEN COLLEGE COMPLETION.|
|Next Article:||GOOD AND BAD REASONS FOR CHANGING A COLLEGE MAJOR: A COMPARISON OF STUDENT AND FACULTY VIEWS.|